Why Your Weight Went Up Overnight (It Isn't Fat)
Why did I gain a kilo overnight?
Because you didn't. Storing a kilogram of body fat requires eating roughly 7,700 kcal more than you burn — about 3,500 kcal per pound — and nobody does that between dinner and breakfast. What the scale picked up was water, food still working its way through you, and waste that hadn't left yet.
The useful thing about this question is that you can settle it with arithmetic rather than faith. Once you've done the sum once, a bad morning reading stops being a verdict and becomes what it is: one noisy measurement.
Do the sum yourself
Fat tissue carries roughly 7,700 kcal per kilogram. So:
A 1 kg overnight gain as fat would need a surplus of about 7,700 kcal — on top of everything you burned that day. That's something like 14 large pizzas' worth of extra energy, eaten and absorbed in twelve hours.
A 2 kg gain would need about 15,000 kcal. Physically, you can't chew and digest that in a day.
Going the other way, a 1 kg overnight loss as fat would need a 7,700 kcal deficit. Most adults don't burn that much in two full days.
Now compare it with the alternative explanation. Water weighs exactly 1 kg per litre. A large glass of water is half a kilo on the scale the moment you drink it. A big dinner weighs what it weighed on the plate, and stays inside you until it doesn't. No mystery required.
What actually moves the number
Salt. Your body keeps the concentration of sodium in your blood within a narrow range. Eat a lot of salt and the quickest fix is to hold on to more water rather than dump the sodium immediately. A single takeaway, a bowl of ramen or a restaurant main can carry a good part of the day's salt on its own — the usual UK adult guideline is about 6 g a day — and the water that comes with it typically clears over the next day or two.
Carbohydrate and glycogen. Carbohydrate is stored in muscle and liver as glycogen, and each gram of glycogen is held alongside roughly 3 g of water. A well-stocked adult carries a few hundred grams of glycogen. Empty those stores with a low-carb stretch or a long run and refill them at a pasta dinner, and the scale can swing by a kilo or more in either direction without a single fat cell changing size. This is also why the first week of almost any diet shows a dramatic drop, and why it never repeats.
Food and waste in transit. Everything you've eaten in the last day or so is still on board, and so is what hasn't been passed yet. A high-fibre day, a change in routine, travel, or simply not having been to the loo yet will all show up.
Heat and breathing. You lose water all night through sweat and every breath you exhale — a few hundred grams is ordinary, more after a hot night or a hard session the evening before.
Short sleep and hard training. A broken night, an unusually late meal, a long flight, or a heavy leg day can all leave you holding more fluid the next morning. Muscle that has been worked hard retains water while it repairs.
Hormonal cycles. For people who menstruate, fluid retention through the second half of the cycle is common and can amount to a kilo or so, resolving on its own. If you're comparing this week to last week, compare the same point in the cycle, or the comparison tells you nothing.
So should I stop weighing myself?
No — the scale is the only direct measurement in the whole business. Everything else you might track is a guess, including the calorie counts. The problem isn't the scale; it's treating one reading as a result rather than one data point in a series.
Daily weighing isn't compulsory either. Twice a week under consistent conditions gives you a usable trend. What matters is that the readings are comparable to each other, not how many of them there are.
How to weigh so the number means something
Hold the conditions still, and the noise shrinks:
Same time of day — first thing in the morning is easiest to repeat.
After using the toilet, before eating or drinking.
Same amount of clothing, or none.
Same scale, same spot on the floor. Hard floor, not carpet.
You aren't trying to find your "true" weight. You're trying to make Tuesday comparable to the Tuesday before.
Why a rolling average is the real signal
Here's the awkward part. A sensible rate of fat loss for most people is somewhere around 0.5 kg a week — which needs a deficit of roughly 500 kcal a day. Day-to-day fluctuation from water and gut contents is routinely larger than that. The signal is smaller than the noise, so any single reading can point the wrong way while everything is going fine.
A rolling seven-day average fixes this by cancelling out the wobbles. Take the mean of the last seven mornings, and compare that number with the same average a week earlier. One salty dinner shifts a seven-day average by about a seventh of its effect. The trend survives; the spike doesn't.
Most tracking apps will draw this line for you. Day So Far does it as a matter of course, because its daily target is derived from the comparison between what you logged and what the trend line actually did — a point reading isn't precise enough to build on.
How many days before a direction is trustworthy?
About a fortnight, as a rough rule, and longer if you're losing slowly.
The faster the real change, the sooner it clears the noise. At around 0.5 kg a week, two weeks of averages usually show an unmistakable slope. At a gentler rate — a quarter of a kilo a week, say — you want three to four weeks before you conclude anything, because one week of unusual salt, travel or cycle timing can swallow the whole difference.
This is also why judging a change of approach after three days is pointless. You haven't got enough data yet to distinguish the change from the weather.
What to do with a bad morning
Write it down and move on. Don't re-weigh in an hour, don't skip lunch to compensate, and don't adjust anything based on it. The reading isn't wrong — your weight genuinely was that this morning. It just wasn't fat, and it won't be there next week.
If the seven-day average has genuinely been flat for three or four weeks, that's a different conversation, and the honest place to start is whether intake is what you think it is. Estimates drift; portions grow quietly. But that's a question for the trend line, not for Thursday.